Robust Integral of NN and Error Sign Control for Nanomanipulation Using AFM

This paper presents a novel control methodology for automatically manipulating nano particles on the substrate by using Atomic Force Microscope (AFM). The interactive forces and dynamics between the tip, particle and substrate are modeled and analyzed including the roughness effect of the substrate. Further, the control signal is designed to consist of the robust integral of a neural network (NN) output plus the sign of the error feedback signal multiplied with an adaptive gain. Using the NN-based adaptive force controller, the task of pushing nano particles is demonstrated in simulation environment. Finally, the asymptotical tracking performance of the closed-loop system, boundedness of the NN weight estimates and applied forces are shown by using the Lyapunov-based stability analysis.